Improvement of virtual metrology performance by removing metrology noises in a training dataset

Dongil Kim, Pilsung Kang, Seung kyung Lee, Seokho Kang, Seungyong Doh, Sungzoon Cho

Research output: Contribution to journalArticle

5 Citations (Scopus)

Abstract

Virtual metrology (VM) has been applied to semiconductor manufacturing processes for the quality management of wafers. However, noises included in training datasets degrade the performance of VM, which is a key obstacle to the application of VM in real-world semiconductor manufacturing processes. In this paper, we develop a VM dataset construction method by identifying and removing noises. We define noises by considering both input and output variables and classify noises into fault detection and classification (FDC) noises and metrology noises, which have abnormal FDC variables and normal metrology variables, and normal FDC variables and abnormal metrology variables, respectively. We propose the construction of a VM training dataset including FDC noises and excluding metrology noises. By employing novelty detection methods, the normal/abnormal regions of FDC variables are identified. In experiments conducted on a real-world photolithography (photo) data, VM models trained with the dataset constructed by the proposed method showed the best accuracy and the most robustness.

Original languageEnglish
Pages (from-to)173-189
Number of pages17
JournalPattern Analysis and Applications
Volume18
Issue number1
DOIs
Publication statusPublished - 2014 Jan 1
Externally publishedYes

Fingerprint

Fault detection
Semiconductor materials
Quality management
Photolithography
Experiments

Keywords

  • Noise identification and removal
  • Novelty detection
  • Semiconductor manufacturing
  • Virtual metrology

ASJC Scopus subject areas

  • Computer Vision and Pattern Recognition
  • Artificial Intelligence

Cite this

Improvement of virtual metrology performance by removing metrology noises in a training dataset. / Kim, Dongil; Kang, Pilsung; Lee, Seung kyung; Kang, Seokho; Doh, Seungyong; Cho, Sungzoon.

In: Pattern Analysis and Applications, Vol. 18, No. 1, 01.01.2014, p. 173-189.

Research output: Contribution to journalArticle

Kim, Dongil ; Kang, Pilsung ; Lee, Seung kyung ; Kang, Seokho ; Doh, Seungyong ; Cho, Sungzoon. / Improvement of virtual metrology performance by removing metrology noises in a training dataset. In: Pattern Analysis and Applications. 2014 ; Vol. 18, No. 1. pp. 173-189.
@article{1cd8bfcded7042d5846ff0db9646208f,
title = "Improvement of virtual metrology performance by removing metrology noises in a training dataset",
abstract = "Virtual metrology (VM) has been applied to semiconductor manufacturing processes for the quality management of wafers. However, noises included in training datasets degrade the performance of VM, which is a key obstacle to the application of VM in real-world semiconductor manufacturing processes. In this paper, we develop a VM dataset construction method by identifying and removing noises. We define noises by considering both input and output variables and classify noises into fault detection and classification (FDC) noises and metrology noises, which have abnormal FDC variables and normal metrology variables, and normal FDC variables and abnormal metrology variables, respectively. We propose the construction of a VM training dataset including FDC noises and excluding metrology noises. By employing novelty detection methods, the normal/abnormal regions of FDC variables are identified. In experiments conducted on a real-world photolithography (photo) data, VM models trained with the dataset constructed by the proposed method showed the best accuracy and the most robustness.",
keywords = "Noise identification and removal, Novelty detection, Semiconductor manufacturing, Virtual metrology",
author = "Dongil Kim and Pilsung Kang and Lee, {Seung kyung} and Seokho Kang and Seungyong Doh and Sungzoon Cho",
year = "2014",
month = "1",
day = "1",
doi = "10.1007/s10044-013-0363-5",
language = "English",
volume = "18",
pages = "173--189",
journal = "Pattern Analysis and Applications",
issn = "1433-7541",
publisher = "Springer London",
number = "1",

}

TY - JOUR

T1 - Improvement of virtual metrology performance by removing metrology noises in a training dataset

AU - Kim, Dongil

AU - Kang, Pilsung

AU - Lee, Seung kyung

AU - Kang, Seokho

AU - Doh, Seungyong

AU - Cho, Sungzoon

PY - 2014/1/1

Y1 - 2014/1/1

N2 - Virtual metrology (VM) has been applied to semiconductor manufacturing processes for the quality management of wafers. However, noises included in training datasets degrade the performance of VM, which is a key obstacle to the application of VM in real-world semiconductor manufacturing processes. In this paper, we develop a VM dataset construction method by identifying and removing noises. We define noises by considering both input and output variables and classify noises into fault detection and classification (FDC) noises and metrology noises, which have abnormal FDC variables and normal metrology variables, and normal FDC variables and abnormal metrology variables, respectively. We propose the construction of a VM training dataset including FDC noises and excluding metrology noises. By employing novelty detection methods, the normal/abnormal regions of FDC variables are identified. In experiments conducted on a real-world photolithography (photo) data, VM models trained with the dataset constructed by the proposed method showed the best accuracy and the most robustness.

AB - Virtual metrology (VM) has been applied to semiconductor manufacturing processes for the quality management of wafers. However, noises included in training datasets degrade the performance of VM, which is a key obstacle to the application of VM in real-world semiconductor manufacturing processes. In this paper, we develop a VM dataset construction method by identifying and removing noises. We define noises by considering both input and output variables and classify noises into fault detection and classification (FDC) noises and metrology noises, which have abnormal FDC variables and normal metrology variables, and normal FDC variables and abnormal metrology variables, respectively. We propose the construction of a VM training dataset including FDC noises and excluding metrology noises. By employing novelty detection methods, the normal/abnormal regions of FDC variables are identified. In experiments conducted on a real-world photolithography (photo) data, VM models trained with the dataset constructed by the proposed method showed the best accuracy and the most robustness.

KW - Noise identification and removal

KW - Novelty detection

KW - Semiconductor manufacturing

KW - Virtual metrology

UR - http://www.scopus.com/inward/record.url?scp=84892970913&partnerID=8YFLogxK

UR - http://www.scopus.com/inward/citedby.url?scp=84892970913&partnerID=8YFLogxK

U2 - 10.1007/s10044-013-0363-5

DO - 10.1007/s10044-013-0363-5

M3 - Article

AN - SCOPUS:84892970913

VL - 18

SP - 173

EP - 189

JO - Pattern Analysis and Applications

JF - Pattern Analysis and Applications

SN - 1433-7541

IS - 1

ER -